Export of Dissolved Methane and Carbon Dioxide with Effluents from

Article pubs.acs.org/est

Export of Dissolved Methane and Carbon Dioxide with Effluents from Municipal Wastewater Treatment Plants Zeyad Alshboul,*,† Jorge Encinas-Fernández,‡ Hilmar Hofmann,‡ and Andreas Lorke† †

University of Koblenz−Landau, Institute for Environmental Sciences, Fortstr. 7, 76829 Landau, Germany University of Konstanz, Limnological Institute, Environmental Physics, Mainaustr. 252, D-78464 Konstanz, Germany

‡

S Supporting Information *

ABSTRACT: Inland waters play an important role for regional and global scale carbon cycling and are significant sources of the atmospheric greenhouse gases methane (CH4) and carbon dioxide (CO2). Although most studies considered the input of terrestrially derived organic and inorganic carbon as the main sources for these emissions, anthropogenic sources have rarely been investigated. Municipal wastewater treatment plants (WWTPs) could be additional sources of carbon by discharging the treated wastewater into the surrounding aquatic ecosystems. Here we analyze seasonally resolved measurements of dissolved CH4 and CO2 concentrations in effluents and receiving streams at nine WWTPs in Germany. We found that effluent addition significantly altered the physicochemical properties of the streamwater. Downstream of the WWTPs, the concentrations of dissolved CH4 and CO2 were enhanced and the atmospheric fluxes of both gases increased by a factor of 1.2 and 8.6, respectively. The CH4 exported with discharged effluent, however, accounted for only a negligible fraction (0.02%) of the estimated total CH4 emissions during the treatment process. The CH4 concentration in the effluent water was linearly related to the organic load of the wastewater, which can provide an empirical basis for future attempts to add WWTPs inputs to regionalscale models for inland water−carbon fluxes. waters.10 Current global estimates of the magnitude of this perturbation are based, however, on the assumption that the anthropogenic sewage enters the inland waters as untreated wastewater, i.e. in organic form.10,13 This is a rather unrealistic assumption in many parts of the world, where ∼98% of the organic C is removed from municipal wastewater during treatment processes prior to its discharge into rivers and streams.14 Wastewater treatment plants (WWTPs) have been demonstrated to be relatively strong, yet poorly constrained, sources of atmospheric CH4.15−17 Anaerobic treatment with CH4 recovery is widely used for energy production at WWTPs. Theoretical considerations suggested that anaerobic treatment can lead to large emission rates and large export rates of dissolved CH4 with effluents.18 Direct measurements at one WWTP supported this suggestion.19 However, systematic studies of the amounts of CH4 and CO2 discharged from different WWTPs effluents are not available. In this study, we quantified the amounts of CH4 and CO2 that are discharged with treated effluents from municipal WWTPs into aquatic ecosystems. We examined the magnitude and variability of the concentrations and atmospheric emissions

1. INTRODUCTION Streams and rivers are essential components of regional and global carbon cycles. They receive large amounts of organic and inorganic carbon (C) from their catchments, a large fraction of which is returned to the atmosphere as carbon dioxide (CO2) and methane (CH4).1−3 A relatively small fraction of the total C is emitted as CH4 (2−4% of the mass of CO2 as a global average).2,4 However, due to the 34-fold global warming potential of atmospheric CH4 in comparison to CO2 over 100 years,5 the total emission of both gases are of comparable importance in terms of their contributions to radiative forcing. Because the majority of the C emitted from rivers and streams was derived from atmospheric CO2 by terrestrial primary production, its return to the atmosphere as CH4 results in a net increase of total atmospheric greenhouse gas (GHG) strength. Current research has been aiming to improve the knowledge on factors and processes that control the terrestrial-aquatic coupling of carbon fluxes from catchment to global scales.6−9 There is a strong evidence that the input, transformation and evasion of C in inland waters are affected by anthropogenic activities,10 e.g., land-use changes in the catchment increase the input of C into rivers11 and sediment trapping by dams can cause strongly enhanced CH4 emission rates.12 Besides soil-derived organic and inorganic C and inorganic C from chemical weathering, sewage inputs have also been considered as an additional anthropogenic C input to inland © XXXX American Chemical Society

Received: October 7, 2015 Revised: April 18, 2016 Accepted: May 10, 2016

A

DOI: 10.1021/acs.est.5b04923 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

sampling sites. In addition to monthly measurements, the shortterm variability of CH4 and CO2 dynamics was measured in the effluent and stream waters of selected WWTPs using continuous measurements over several days. 2.2. Physicochemical Measurements and Analyses. In situ water temperature (°C), dissolved oxygen concentration DO2 (μM), pH, specific conductivity (μS cm−1) and redox potential (mV) were measured approximately 10 cm below the water surface using water quality sensors (see SI section S1.1 for further details). Water samples were collected for laboratory analyses of dissolved inorganic carbon (DIC), total and dissolved organic carbon (TOC and DOC) using a Total Carbon Analyzer (Multi N/C 2100, Analytik Jena, Germany). For DOC and DIC, the samples were filtered using 0.45 μm hydrophilic polypropylene membranes. The procedures for sampling, preparation and storage were conducted according to the German standard methods for the examination of water, wastewater and sludge (after DIN 38406; E5.1). Additional physicochemical parameters of the wastewater inflow and effluents were obtained from the operational monitoring program of the WWTPs (Table S2, SI). The parameters include water temperature, pH and specific conductivity of accumulated samples over 24 h. Biweekly measurements and monthly mean values of biological oxygen demand (BOD5) and chemical oxygen demand (COD5) were obtained for the WWTPs (except for Edenkoben where seasonal mean values were available). The monthly mean volume of inflowing wastewater and effluent discharge were obtained from continuous monitoring of the WWTPs. The removal efficiency (%) of BOD5 (BODe) and COD5 (CODe) was calculated from the normalized difference of both rates in the inflowing (Inf.) and effluent (Eff.) waters of the WWTPs according to Spellman:20

of CH4 and CO2 at the effluents of nine WWTPs in Southwest Germany over a period of 12 months. The results are discussed with respect to the contribution of the effluent discharge to (i) the total emission rates of the WWTPs and (ii) the CO2 and CH4 dynamics in the effluent-receiving streams.

2. MATERIALS AND METHODS 2.1. Study Sites and Sampling Overview. Measurements were performed at nine municipal WWTPs in Southwest Germany (Annweiler, Bellheim, Edenkoben, Germersheim, Hochstadt, Landau, Neustadt, Offenbach and Ruelzheim) (Figure 1), which serve between 5.6 and 90 thousand

⎛ Inf . −Eff . ⎞ removal efficiency = ⎜ ⎟ × 100% ⎝ Inf . ⎠

Figure 1. Location of the studied WWTPs within the stream network of the state of Rhineland Palatinate, Germany. The region’s WWTPs and the studied WWTPs are denoted by the red and green dots, respectively. The size of the region’s WWTPs (in population equivalent PE) are denoted by the size of the red dots and ranged between 20 and 100 thousand PE. The studied WWTPs are labeled as follow: Annweiler (A), Bellheim (B), Edenkoben (E), Hochstadt (H), Germersheim (G), Landau (L), Neustadt (N), Offenbach (O) and Ruelzheim (R).

(1)

Water discharge at the upstream sampling site in the receiving water systems (Qup., m3 d−1) was estimated from the measured specific conductivity using a mass-balance approach (dilution gauging method):21 ⎛γ − γ ⎞ down ⎟ Q up. = Q eff. × ⎜⎜ eff. ⎟ ⎝ γdown − γup ⎠

population equivalents (Table 1). The WWTPs apply primary (bar screen and primary settler) and secondary (aerobic/ anaerobic digester and secondary clarifier) treatment. The treatment technology was consistent among the WWTPs except for two plants (Bellheim and Hochstadt) which were not equipped with anaerobic sludge treatment for energy production (Table 1). We measured the partial pressure of dissolved carbon dioxide (pCO2) and methane (pCH4), the diffusive air−water flux of CO2, and additional physicochemical parameters on a monthly basis between March 2014 and March 2015. All measurements were performed in the effluent waters and up- and downstream of the effluent discharge location in the receiving streams. The downstream sampling sites were located where effluent and streamwater were well mixed. The mixing points were specified based on negligible transversal gradients in specific conductivity and varied between 30 and 100 m distance from the effluent discharge. The longitudinal decay of the dissolved gas concentration (CH4 and CO2) downstream of the WWTP was measured exemplarily at two

(2)

where Qeff. (m3 d−1) is the effluent’s discharge, γeff, γdown and γup are the specific conductivities (μS cm−1) of effluent, downstream and upstream water, respectively. 2.3. Concentrations and Fluxes of CH4 and CO2. pCH4 and pCO2 in water samples were measured using the headspace method.22 The headspace was created in a glass reagent bottle and the gas partial pressure was measured on-site in a closed gas loop with an ultraportable greenhouse gas analyzer (UGGA, Los Gatos Research Inc., USA) (Figure S1).23 The gas analyzer was factory calibrated in the range of 0.01−100 ppm for CH4 (repeatability 0.6 ppb) and in the range of 1−20 000 ppm (repeatability < 100 ppb) for CO2. The concentration of dissolved CH4 and CO2 (μM) was obtained by multiplication of the calculated partial pressures (μatm) of both gases with the solubility coefficient (mol l−1 atm−1)24 (see SI section S1.2). The annual mean mass of dissolved CH4 (ECH4dis., g CH4 yr−1) and CO2 (ECO2dis., g CO2 yr−1) exported with the effluent water was calculated as the mean product of dissolved B


Article

Environmental Science & Technology Table 1. Characteristics of the WWTPa Ruelzheim 3

PE (×10 ) RWS Qeff. (l s−1) Qup/Qeff. DCO2eff. (μM) DCH4eff. (μM) BODe. (%) CODe. (%) TWO (×105) (kg BOD yr−1) EE (×105) (kg CH4 yr−1) ECO2cap. (×103) (g CO2 yr−1 capita−1) ECH4cap. (g CH4 yr−1 capita−1)

Offenbach

Neustadt 51 Speyrbach 122 ± 37 0.8 ± 0.6 707 ± 105 0.18 ± 0.07 99.60 95.60 14.4

90 Queich 170

402 ± 142 0.19 ± 10.17 97 92.40 7.07

6 Queich 23 ± 3.5 0.2 ± 0.3 655 ± 194 0.3 ± 0.7 99 45 1.7

2.12 0.6 ± 0.06

0.5 3.6 ± 1.5

0.1 ± 0.1

0.6 ± 1.3

41.5 Klingbach ∼46

Landau

Hochstadt

Germersheim

Edenkoben

Bellheim

Annweiler

14.3 Muhlbach 38 ± 8 1 ± 0.9 444 ± 97 6.5 ± 5 90 95.90 4.05

17 Queich 54 ± 8 4.8 ± 4.7 475 ± 93 0.2 ± 0.07 98.90 95.10 4.81

35 Queich 78.7

688 ± 107 0.75 ± 0.45 99.30 96.50 25.5

5.6 Hainbach 17.3 ± 2.3 1.4 ± 1.4 662 ± 178 0.76 ± 0.6 99.40 97.60 1.58

596 ± 110 0.28 ± 0.12 99.30 97.20 9.9

38 Triefenbach 56 ± 7 0.67 ± 0.7 575 ± 87 1.4 ± 0.6 98.70 95 10.7

4.33 2.4 ± 0.8

7.64 1.8 ± 0.3

4.75 3.8 ± 2.2

2.97 1.9 ± 0.3

3.22 1.2 ± 0.2

1.21 1.9 ± 0.5

1.44 2.2 ± 0.6

0.25 ± 0.1

0.7 ± 0.4

1.5 ± 1.1

0.3 ± 0.14

0.9 ± 0.4

10 ± 7.8

0.33 ± 0.15

PE are population equivalents of the design capacity. RWS is the name of the receiving stream. Qeff. (l s−1) is the annual mean discharge of effluent water and Qup./Qeff. is the ratio of the annual mean stream discharge at the upstream sampling site to the effluent discharge. DCH4eff. and DCO2eff. are the mean dissolved concentrations of CH4 and CO2 in the effluent (μM). BODe. and CODe. are the removal efficiency of BOD5 and COD5 (%), respectively. TWO is the total organic in wastewater (kg BOD yr−1) and EE is the methane emission estimation in (kg CH4 yr−1) by eq S5. ECO2cap. (g CO2 yr−1 capita−1) and ECH4cap. (g CH4 yr−1 capita−1) are exported CO2 and CH4 with effluents per capita. Except for the facilities in Bellheim and Hochstadt, all treatment plants had an energy recovery system based on anaerobic digestion. Where applicable, the data are provided as mean ± SD. a

dissolved CH4 concentration in their effluents during an initial survey. During the first campaign, dissolved CH4 concentration was measured continuously using two submersible sensors (HydroC, Contros, Germany) in the effluent and upstream water of Neustadt WWTP between July 22 and August 1, 2014. Continuous measurements of dissolved oxygen concentration and water temperature were simultaneously obtained from temperature-oxygen loggers (MiniDOT, Precision Measurement Engineering, USA), which were deployed directly in the corresponding water system. During the second campaign pCH4 and pCO2 were measured in the effluent water of Bellheim WWTP for 5 days (from October 28 to November 3, 2014). Both gases were measured with a UGGA gas analyzer in a closed gas loop connected to a flow-through membrane contactor (MiniModule, Liqui-Cel, USA).28 The effluent water was pumped through the contactor at a flow rate of ∼1−3 L min−1. The third campaign consisted of a long-term deployment of a floating chamber to obtain continuous measurements of pCH4 and pCO2 in the chamber head space, which was assumed to be in equilibrium with the surface water. The chambers were installed December 8−11, 2014 in the effluent canal of Bellheim WWTP. The pCH4 and pCO2 in the chamber headspace were measured with the UGGA gas analyzer in a closed gas loop. The response time of the chamber depends on the local gas exchange velocity and on gas solubility, and it was estimated to be in the order of ∼0.23 h for CO2 and ∼3.8 h for CH4. The three different methods applied during the continuous measurement campaigns were chosen based on instrument availability. All three methods have shown good agreement with the headspace sampling technique used during the monthly campaigns over a wide range of dissolved gas partial pressures and water chemical properties, including high concentrations of dissolved organic carbon.23,25,29 2.5. Downstream Gas Loss Rate. The residence time of the exported CH4 and CO2 in the receiving streams was estimated for the WWTP in Landau and Edenkoben. By assuming first-order decay of an initial CH 4 and CO 2

CH4 and CO2 concentration and effluent discharge. Per capita exported CH4 (ECH4cap., g CH4 capita−1 yr−1) and CO2 (ECO2cap., g CO2 capita−1 yr−1) of each WWTP was obtained by dividing the exported rates of both gases by the number of served population. Atmospheric CO2 fluxes (FCO2, mg CO2 m−2 d−1) across the air−water interface were determined using floating chambers (surface area, 0.078 m2; chamber volume, 7.66 l). The floating chambers were equipped with an internal CO2 logger (ELG, Senseair, Sweden) and were deployed as triplicates for approximately 30 min. The CO2 logger consisted of a nondispersive infrared CO2 sensor with a measurement range of 0−5000 ppm, and measured simultaneously temperature and relative humidity (RH %). The nominal operating temperature of the sensor is 0−50 °C, with a full function at high humidity (noncondensing conditions). The loggers were operated with 9 VDC batteries and calibrated by the manufacturer. More information on this particular type of chamber can be found elsewhere.25,26 The loggers were calibrated before each measurement in the range of about 400−5000 ppm by following the procedure described in Bastviken et al. 2015.25 The CH4 flux (FCH4, mg CH4 m−2 d−1) was estimated from the gas exchange velocity observed in the chamber deployments and measured pCH4 (see SI section S1.3). Bubble Fluxes. An automated bubble trap (ABT, Senect, Germany) was used to measure the volume of methane emitted as bubbles from the effluent systems. The ABT is similar to that described by Maeck et al.27 and consists of an inverted polypropylene funnel (diameter: 1 m), a cylindrical capture trap and a custom-made electronic unit equipped with a differential pressure sensor. ABT deployment was conducted in the effluent of Bellheim WWTP from March 12 to April 23, 2015. 2.4. Continuous Measurement for pCH4 and pCO2. The short-term (minutes−days) variability of pCH4 and pCO2 in the stream and effluent water of two selected WWTPs (Bellheim and Neustadt) was measured during three different campaigns. These WWTPs were selected based on the observed maximum (Bellheim) and minimum (Neustadt) C


Article


Figure 2. Comparison of various physicochemical properties observed in the WWTP effluents (Effl., black) and in streamwater up- (Up, green) and down- (Down, blue) stream of the effluent discharge in the receiving stream. The diamond-shaped boxes show the 25 and 75 percentile range. The whiskers extend this range by a factor of 1.5 and data beyond this extended range are not shown. The median and mean values are marked by horizontal lines and open squares, respectively: (a) temperature (T), (b) dissolved oxygen concentration (DO2), (c) dissolved organic carbon (DOC), (d) dissolved inorganic carbon (DIC), (e) dissolved CO2 concentration (DCO2), (f) dissolved CH4 concentration (DCH4), (g) CO2 flux (FCO2) and (h) CH4 flux (FCH4).

The Shapiro−Wilk normality test was performed prior to parametric statistical analysis. Analysis of covariance (ANOVA) was used to test the temporal and spatial variability among the sampling sites and monthly measurements. All statistical analyses were performed with the software OriginPro 9.1.

concentration at the downstream sampling site, where effluent and streamwater are well mixed, the longitudinal gas loss rate γ (m−1) can be estimated as30

( )

ln γ=−

C(d) C0

d

(3)

3. RESULTS 3.1. Influence of Effluent on Physicochemical Properties of Streamwater. The nine WWTPs received wastewater from 298 thousand population equivalents and discharged a total amount of 20 million m3 of effluent water annually. The average amount of treated wastewater was 67 m3 yr−1 capita−1. The effluent was diluted by streamwater in the receiving aquatic system, and the average ratio of upstream and effluent discharge rates ranged between 0.2 and 5 (Table 1). However, the effluent discharge exceeded the upstream discharge at 2 out of 6 sampling sites. The effluent additions had significant effects on all measured physicochemical properties of the stream waters. On average, the downstream waters had higher temperature (+1.5 °C), DOC (+0.46 mg L−1) and DIC (+4.3 mg L−1) concentration, and lower DO2 concentration (−18.75 μM) in comparison to the upstream sites (Figure 2). The mean pH at the downstream sites (7.6 ± 0.5) was lower than at the upstream sites (7.9 ± 0.3) (Figure S4; Table S3 and Table S4), which could be associated with the increase in DIC.

where C(d) is the gas concentration measured at the distance d (m) from the downstream sampling point and C0 is the concentration measured at the downstream sampling point. In addition to C0, C(d) was measured at a distance of 50 m down from C0 during the monthly measurements at Landau and Edenkoben WWTPs. We used γ to estimate the longitudinal decay length at which the concentration of both gases has decreased to the respective upstream concentration. 2.6. Estimation of WWTP CH4 Emissions. To quantify the relative importance of CH4 exported in effluents to the total CH4 footprint of the WWTPs, the emissions during wastewater treatment were estimated according to the method proposed by the IPCC 2006, Tier 1 method,31 with default values for the activity parameters and the emission factor. The emission factor relates the in-plant methane production rate to the BOD of the treated wastewater (see SI section S1.4, for further details). 2.7. Statistical Analyses. A significance level of p = 0.05 was considered for all statistical tests, unless stated otherwise. D


Article


Figure 3. Relationships between (a) excess DO2 and excess DCO2, (b) excess DO2 and excess DCH4, (c) excess DCO2 and water temperature (T) and (d) excess DCH4 and water temperature. Data from all nine WWTPs are shown; the colors indicate the sampling location: effluent (black), downstream (blue) and upstream (green).

Dissolved CH4 and CO2 concentrations in the stream and effluent waters showed seasonal variations, with maximum values during summer (June and July, Figure S5 and Figure S6). The excess DCO2, i.e., difference between water concentrations and atmospheric equilibrium dissolved concentrations, was positively correlated to water temperature at the upstream sampling sites but negatively correlated to the excess DO2 at both downstream and upstream sampling sites (Figure 3a,c). There were no strong relationships between DCH4 and DCO2 with redox potential, TOC, DOC and DIC (some of these correlations were significant for p < 0.05, but had a Pearson correlation coefficient r < 0.35, which indicates very low predictive power). Excess DCH4 was positively correlated to temperature at the downstream and upstream sites and negatively correlated to excess DO2 at most of the sampling sites (Figure 3b,d). Significant positive correlations were observed between excess DCH4 and DCO2 at the upstream sampling sites. 3.3. Short-Term Variations, Ebullition and Downstream Loss Rates. Short-term variations during four to nine days of continuous DCH4 measurements in the effluent and at the upstream sampling sites of two selected WWTPs were smaller or comparable in magnitude to the seasonal variations observed during the monthly sampling campaigns (Figure 4). The measurements conducted in July at Neustadt WWTP indicated that upstream CH4 concentrations exceeded those in the effluent (Figure S2). In contrast to the upstream sites, the effluent showed a pronounced diurnal pattern with elevated CH4 and low O2 concentrations during the night (see SI section S2.1.1). Less short-term variability and no diurnal DCH4eff variations were observed during the two measurement periods at Bellheim WWTP in October and December (Figure 4b; see SI section S2.1.2). Emissions via gas bubbles were below the detection limit of the ABT ( 0.1; Figure 5). By excluding the measurements made at Bellheim WWTP from the analysis, the correlation became significant (p < 0.05, r2 = 0.8) and the

DCH4eff. = 2.37 × 10−5COD5 − 0.09

(4)

The annual mass of dissolved CO2 exported with the WWTP effluents (ECO2dis.) varied strongly among the nine WWTPs and was on average of 60 ± 50 Mg CO2 yr−1. Mean exported CO2 (ECO2cap.) of the nine WWTPs was 2.8 ± 3 kg CO2 capita−1 yr−1 (Figure 6a, Table 1). The annual mass of CH4 exported with the effluents of the nine WWTPs (ECH4dis.), was

Figure 6. Per capita export of (a) CO2 and (b) CH4 via the effluent of the different WWTPs. The diamond-shaped boxes show the 25 and 75 percentile range. The whiskers extend this range by a factor of 1.5 and data beyond this extended range are not shown. The median and mean values are marked by horizontal lines and open squares, respectively. F


Article

Environmental Science & Technology on average of 30.9 ± 40.7 kg CH4 yr−1 corresponding to 1.6 ± 2.7 g CH4 capita−1 yr−1 (Figure 6b, Table 1). The mean CH4 emission during the treatment process of the nine WWTPs was estimated to be 266 ± 227 Mg CH4 yr−1, and ranged from 47.5 to 764 Mg CH4 yr−1 (Table 1). Hence, 0.02% (range = 0.002−0.1%) of the total estimated CH4 emission during the treatment process was exported in dissolved form with the effluent.

was stored in a storage tank prior to discharge (SI section 2.1.2). The hydraulic design of the tank outflow was less turbulent, resulting in less degassing. Storage tanks also extend the water retention time, which has been shown to be a key factor modulating CH4 production in sewers.34 The exceptionally high specific export rate at Bellheim WWTP could also be related to the fact that this treatment plant receives ∼20% of its wastewater from a brewery. However, we could not identify specific differences in the physicochemical properties of the wastewater in Bellheim in comparison to those measured in the other WWTP. Further studies are needed to investigate the exact factors and processes that cause the high discharge of CH4 from this plant. Variations of dissolved CH4 concentrations in the effluents of the remaining WWTPs could partially be explained by a correlation with the COD5 of the inflow wastewater among the plants (Figure 5). This indicates that the quality of the wastewater has some effect on the dissolved CH4 in the effluent. The diurnal variations of dissolved CH4 concentration observed in the effluents are in accordance with diurnal variations of CH4 emissions from WWTPs during different treatment steps.19 Comparative studies including three treatment plants in The Netherlands pointed toward higher emission factors from WWTPs, which apply anaerobic sludge treatment for biogas production and energy recovery.32 We did not find an indication for lower CH4 discharge rates in two of the nine WWTPs, which had no anaerobic sludge digester (Bellheim and Hochstadt, Table 1). In contrast to CH4, significant amounts of inorganic carbon were exported from the WWTPs as DIC and excess CO2 in the effluents. The mole fraction of BOD of the wastewater, which was exported as DIC with the effluent water, was 3.3−14.4% indicating that export by effluent water might be a more significant component of the WWTPs carbon budget. 4.2. CH4 and CO2 Imported by the Receiving Streams. In contrast to the negligible relative importance for the treatment plant CH4 and C budgets, the effluent water significantly affected downstream CH4 and CO2 concentrations and emission rates in the receiving streams. Although CO2 concentration and fluxes increased consistently several-fold in comparison to the upstream sampling locations, the average increase of CH4 concentrations and fluxes was more variable among the WWTPs. Most of the downstream samples (89%) had higher CH4 concentration than the upstream samples. This concentration increase could be partially explained by the addition of effluent water, which had higher CH4 concentrations than the upstream samples in only 53% of the samples. The unexplained concentration increase at some of the sites suggests enhanced CH4 production in the stream section between the up and downstream sampling sites, which could potentially be attributed to the additional organic carbon load in the effluent water. Enhanced emission rates of nitrous oxide (N2O) have been observed downstream of a WWTP in a large river,35 but to the best of our knowledge no study has so far quantified the effect of WWTPs effluent on downstream emissions of CH4 and CO2. The median value of the CH4 fluxes measured at the downstream sampling sites (13 mg CH4 m−2 d−1) was comparable to the mean flux assigned to streams in the temperate zone in a global estimation of freshwater CH4 emissions (13.3 mg CH4 m−2 d−1).4 Measured fluxes of CO2 were smaller than the global mean value for streams in the temperate zone (2.6 × 104 mg CO2 m−2 d−1).3 Hence, the

4. DISCUSSION 4.1. CH4 and CO2 Exported from the Treatment Plants. The export of dissolved CH4 with effluent discharge accounts for only a negligible fraction of the estimated total CH4 emissions of the WWTPs. This general finding is neither affected by the high variability of export rates among the WWTPs nor by the high uncertainty (50%), which is associated with the emission factor with country-default parameters.17 Extensive flux measurements showed that the actual CH4 emissions from WWTPs may substantially deviate from the value obtained from the IPCC Tier 1 method.32 The linear correlation observed between CODinf. and DCH4eff., however, can be considered as a confirmation of the linear relationship between the total organic carbon in wastewater and the CH4 production during the treatment processes (eq S5). The negligible fraction (0.02%) of DCH4eff. relative to the in-plant CH 4 emission provides experimental support for the assumption made in existing approaches for estimating plantscale CH4 emission rates based on point-wise measurements and mass balances, where the dissolved CH4 exported in effluents is neglected.16,19,33 CH4 production during wastewater treatment depends mainly on the quantity of degradable organic material (measured as BOD5 and COD5) in wastewater, the temperature, the type of treatment, and particularly on the extent to which the wastewater is treated anaerobically.31 Plant-scale emission measurements have shown that anaerobic sludge treatment comprises a major in-plant source for atmospheric CH4 emissions.19,33 It has been suggested that a large fraction of the CH4 produced during anaerobic digestion will be in dissolved form.18 The CH4 concentrations measured in the effluents of the present study, however, were more than 2 orders of magnitude smaller than those estimated by Liu et al.18 We attribute this large difference to the fact that the authors estimated the dissolved fraction of CH4 by assuming a partial pressure of 1 atm (pure CH4 atmosphere). This is a very unlikely situation due to the very low solubility of CH4 in comparison to, e.g., CO2. The CH4 partial pressure in the effluent water was always below 10.8 × 103 μatm in our observations. However, it cannot be excluded that the majority of formerly dissolved CH4 was emitted within the WWTP, before the effluents reached our sampling sites. It can be expected that the in-plant emissions from the effluent water prior to our sampling were strongly affected by the hydraulic design, flow velocity and aeration conditions in the particular WWTP (Figure S7). Variations in constructional features provide a potential explanation for the large differences in the specific amounts of CH4 exported with the effluents from the WWTPs (Figure 6). For instance, the direct discharge of the effluent water into the receiving stream (e.g., at Neustadt WWTP) at a moderate or high flow velocity caused enhanced degassing of CH4. This was apparent from the low DCH4eff. values at Neustadt WWTP (SI section 2.1.1). Much higher DCH4eff. was observed at Bellheim WWTP, where the effluent G


Article

Environmental Science & Technology fluxes of both greenhouse gases measured at the downstream sampling sites were not exceptionally high in comparison to global figures, but were significantly higher than the fluxes measured upstream of the WWTPs at most study sites. Despite the significant effect of WWTPs effluent on the downstream concentration of dissolved gases, the gas concentration in the streamwater measured at the up- and downstream sampling sites had similar characteristics. Surface water excess CH4 and CO2 were positively correlated at most down- and upstream sampling sites. Such correlation has also been observed in the Amazon river,36 in African inland waters,37 and in boreal lakes.22 The observed negative correlation between both DCH4 and DCO2 with DO2 is an indication of the importance of stream metabolism for the dynamics of both gases. The flow length over which the CH4 and CO2 concentrations were affected by the effluent discharge was estimated at one WWTP, resulting in a decay length of up to 5.5 and 11.9 km for DCH4 and DCO2, respectively. However, applying the gas loss rate measured at the Landau WWTP to the high CH4 concentration measured in Bellheim, results in a decay length of 9.6 km. To assess the potential importance of WWTP effluent on GHG concentration and fluxes in rivers and streams, we divided the total length of rivers and streams in Germany (127 × 103 km)38 by the total number of WWTPs (10 × 103),39 which results in an average density of one treatment plant every 12.7 km. Because the decay length can be expected to depend on the flow velocity and on the gas exchange velocity in the receiving aquatic system, the limited measurements obtained in the present study may not be representative. Nevertheless, the close agreement between the decay length and the larger-scale WWTP density demonstrates the potential extent, to which CO2 and CH4 emissions from streams and rivers are affected by effluents. Current approaches for connecting terrestrial and aquatic carbon budgets at the landscape scale have mainly focused on the input of organic and inorganic carbon from terrestrial ecosystems as the major controlling factors.7−9 Our findings suggest that WWTPs can be important point sources for CH4 and CO2 of receiving water systems at regional scales, which should be taken into consideration in future assessments. From this perspective, an improved understanding of the operational and constructional parameters of WWTPs, which eventually control the strongly variable plant-specific export rates of CH4, is required. The observed correlation between CH4eff. and COD5 of the untreated wastewater may provide an empirical basis for future attempts to add WWTPs inputs to regionalscale models for inland water−carbon fluxes.

■

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge the financial support provided by the German Academic Exchange Service (DAAD) - Sustainable water management Program (NAWAM, Grant no. A/12/ 91768) and by the Germany Research Foundation (Grant number: LO 1150/9-1). Authors Jorge Encinas-Fernández and Hilmar Hofmann received funding from Young Scholar Fund at the University of Konstanz (YSF-DFG, 419-14). We also thank the staff and operators of the wastewater treatment plants for their support and for providing data.

■

(1) Raymond, P. A.; Hartmann, J.; Lauerwald, R.; Sobek, S.; McDonald, C.; Hoover, M.; Butman, D.; Striegl, R.; Mayorga, E.; Humborg, C.; Kortelainen, P.; Durr, H.; Meybeck, M.; Ciais, P.; Guth, P. Global carbon dioxide emissions from inland waters. Nature 2013, 503 (7476), 355−359. (2) Cole, J. J.; Prairie, Y. T.; Caraco, N. F.; McDowell, W. H.; Tranvik, L. J.; Striegl, R. G.; Duarte, C. M.; Kortelainen, P.; Downing, J. A.; Middelburg, J. J.; Melack, J. Plumbing the Global Carbon Cycle: Integrating Inland Waters into the Terrestrial Carbon Budget. Ecosystems 2007, 10 (1), 172−185. (3) Aufdenkampe, A. K.; Mayorga, E.; Raymond, P. A.; Melack, J. M.; Doney, S. C.; Alin, S. R.; Aalto, R. E.; Yoo, K. Riverine coupling of biogeochemical cycles between land, oceans, and atmosphere. Front. Ecol. Environ. 2011, 9 (1), 53−60. (4) Bastviken, D.; Tranvik, L. J.; Downing, J. A.; Crill, P. M.; EnrichPrast, A. Freshwater Methane Emissions Offset the Continental Carbon Sink. Science 2011, 331 (6013), 50. (5) IPCC. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, U. K. and New York, NY, 2013; p 1535. (6) Hotchkiss, E. R.; Hall, R. O., Jr.; Sponseller, R. A.; Butman, D.; Klaminder, J.; Laudon, H.; Rosvall, M.; Karlsson, J. Sources of and processes controlling CO2 emissions change with the size of streams and rivers. Nat. Geosci. 2015, 8 (9), 696−699. (7) Teodoru, C. R.; Nyoni, F. C.; Borges, A. V.; Darchambeau, F.; Nyambe, I.; Bouillon, S. Dynamics of greenhouse gases (CO2, CH4, N2O) along the Zambezi River and major tributaries, and their importance in the riverine carbon budget. Biogeosciences 2015, 12 (8), 2431−2453. (8) Lauerwald, R.; Hartmann, J.; Moosdorf, N.; Kempe, S.; Raymond, P. A. What controls the spatial patterns of the riverine carbonate system?  A case study for North America. Chem. Geol. 2013, 337− 338 (0), 114−127. (9) Humborg, C.; Morth, C.-M.; Sundbom, M.; Borg, H.; Blenckner, T.; Giesler, R.; Ittekkot, V. CO2 supersaturation along the aquatic conduit in Swedish watersheds as constrained by terrestrial respiration, aquatic respiration and weathering. Global Change Biol. 2010, 16 (7), 1966−1978. (10) Regnier, P.; Friedlingstein, P.; Ciais, P.; Mackenzie, F. T.; Gruber, N.; Janssens, I. A.; Laruelle, G. G.; Lauerwald, R.; Luyssaert, S.; Andersson, A. J.; Arndt, S.; Arnosti, C.; Borges, A. V.; Dale, A. W.; Gallego-Sala, A.; Godderis, Y.; Goossens, N.; Hartmann, J.; Heinze, C.; Ilyina, T.; Joos, F.; LaRowe, D. E.; Leifeld, J.; Meysman, F. J. R.; Munhoven, G.; Raymond, P. A.; Spahni, R.; Suntharalingam, P.; Thullner, M. Anthropogenic perturbation of the carbon fluxes from land to ocean. Nat. Geosci. 2013, 6 (8), 597−607. (11) Raymond, P. A.; Oh, N.-H.; Turner, R. E.; Broussard, W. Anthropogenically enhanced fluxes of water and carbon from the Mississippi River. Nature 2008, 451 (7177), 449−452. (12) Maeck, A.; Delsontro, T.; McGinnis, D. F.; Fischer, H.; Flury, S.; Schmidt, M.; Fietzek, P.; Lorke, A. Sediment trapping by dams

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b04923. Further details on methods and supplementary data; the date includes monthly variation of measured physical and chemical parameters and an additional discussion of structural features of WWTPs (PDF).

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Z. Alshboul. Email: [email protected]. Tel: +49 (6341) 280-31317. Fax: +49 (6341) 280-31-577. H


Article

Environmental Science & Technology creates methane emission hot spots. Environ. Sci. Technol. 2013, 47 (15), 8130−8137. (13) Ver, L. M. B.; Mackenzie, F. T.; Lerman, A. Biogeochemical responses of the carbon cycle to natural and human perturbations; past, present, and future. Am. J. Sci. 1999, 299 (7−9), 762−801. (14) Chan, Y. J.; Chong, M. F.; Law, C. L.; Hassell, D. G. A review on anaerobic−aerobic treatment of industrial and municipal wastewater. Chem. Eng. J. 2009, 155 (1−2), 1−18. (15) Yoshida, H.; Monster, J.; Scheutz, C. Plant-integrated measurement of greenhouse gas emissions from a municipal wastewater treatment plant. Water Res. 2014, 61, 108−18. (16) Czepiel, P. M.; Crill, P. M.; Harriss, R. C. Methane emissions from municipal wastewater treatment processes. Environ. Sci. Technol. 1993, 27 (12), 2472−2477. (17) Hiller, R. V.; Bretscher, D.; DelSontro, T.; Diem, T.; Eugster, W.; Henneberger, R.; Hobi, S.; Hodson, E.; Imer, D.; Kreuzer, M.; Künzle, T.; Merbold, L.; Niklaus, P. A.; Rihm, B.; Schellenberger, A.; Schroth, M. H.; Schubert, C. J.; Siegrist, H.; Stieger, J.; Buchmann, N.; Brunner, D. Anthropogenic and natural methane fluxes in Switzerland synthesized within a spatially explicit inventory. Biogeosciences 2014, 11 (7), 1941−1959. (18) Liu, Z. H.; Yin, H.; Dang, Z.; Liu, Y. Dissolved methane: a hurdle for anaerobic treatment of municipal wastewater. Environ. Sci. Technol. 2014, 48 (2), 889−90. (19) Daelman, M. R. J.; van Voorthuizen, E. M.; van Dongen, U. G. J. M.; Volcke, E. I. P.; van Loosdrecht, M. C. M. Methane emission during municipal wastewater treatment. Water Res. 2012, 46 (11), 3657−3670. (20) Spellman, F. R. Mathematics manual for water and wastewater treatment plant operators: basic mathematics for water and wastewater operators; CRC Press: Boca Raton, FL, 2014. (21) Rantz, S. E. Measurement and computation of streamflow: Volume 1, Measurement of stage and discharge; Water Supply Paper 2175; U.S. Geological Survey: Washington, DC, 1982. (22) Campeau, A.; Del Giorgio, P. A. Patterns in CH4 and CO2 concentrations across boreal rivers: Major drivers and implications for fluvial greenhouse emissions under climate change scenarios. Glob Chang Biol. 2014, 20 (4), 1075−88. (23) Gonzalez-Valencia, R.; Magana-Rodriguez, F.; Gerardo-Nieto, O.; Sepulveda-Jauregui, A.; Martinez-Cruz, K.; Anthony, K. W.; Baer, D.; Thalasso, F. In Situ Measurement of Dissolved Methane and Carbon Dioxide in Freshwater Ecosystems by Off-Axis Integrated Cavity Output Spectroscopy. Environ. Sci. Technol. 2014, 48 (19), 11421−11428. (24) Goldenfum, J. A. GHG measurement guidelines for freshwater reservoirs. UNESCO/IHA, London, U. K., 2011. (25) Bastviken, D.; Sundgren, I.; Natchimuthu, S.; Reyier, H.; Gålfalk, M. Technical Note: Cost-efficient approaches to measure carbon dioxide (CO2) fluxes and concentrations in terrestrial and aquatic environments using mini loggers. Biogeosciences 2015, 12 (12), 3849−3859. (26) Lorke, A.; Bodmer, P.; Noss, C.; Alshboul, Z.; Koschorreck, M.; Somlai-Haase, C.; Bastviken, D.; Flury, S.; McGinnis, D. F.; Maeck, A.; Müller, D.; Premke, K. Technical note: drifting versus anchored flux chambers for measuring greenhouse gas emissions from running waters. Biogeosciences 2015, 12 (23), 7013−7024. (27) Maeck, A.; Hofmann, H.; Lorke, A. Pumping methane out of aquatic sediments − ebullition forcing mechanisms in an impounded river. Biogeosciences 2014, 11 (11), 2925−2938. (28) Roehm, C. L.; Prairie, Y. T.; del Giorgio, P. A. The pCO2 dynamics in lakes in the boreal region of northern Québec, Canada. Global Biogeochem. Cycles 2009, 23 (3), GB3013. (29) Abril, G.; Bouillon, S.; Darchambeau, F.; Teodoru, C. R.; Marwick, T. R.; Tamooh, F.; Ochieng Omengo, F.; Geeraert, N.; Deirmendjian, L.; Polsenaere, P.; Borges, A. V. Technical Note: Large overestimation of pCO2 calculated from pH and alkalinity in acidic, organic-rich freshwaters. Biogeosciences 2015, 12 (1), 67−78. (30) Crawford, J. T.; Lottig, N. R.; Stanley, E. H.; Walker, J. F.; Hanson, P. C.; Finlay, J. C.; Striegl, R. G. CO2 and CH4 emissions

from streams in a lake-rich landscape: Patterns, controls, and regional significance. Global Biogeochem. Cycles 2014, 28 (3), 197−210. (31) IPCC. IPCC Guidelines for National Greenhouse Gas Inventories; Institute for Global Environmental Strategies (IGES): Japan, 2006; available at: http://www.ipcc-nggip.iges.or.jp/public/ 2006gl/index.html (last accessed May 1, 2015). (32) Daelman, M. R.; van Voorthuizen, E. M.; van Dongen, L. G.; Volcke, E. I.; van Loosdrecht, M. C. Methane and nitrous oxide emissions from municipal wastewater treatment - results from a longterm study. Water Sci. Technol. 2013, 67 (10), 2350−5. (33) Yver Kwok, C. E.; Müller, D.; Caldow, C.; Lebègue, B.; Mønster, J. G.; Rella, C. W.; Scheutz, C.; Schmidt, M.; Ramonet, M.; Warneke, T.; Broquet, G.; Ciais, P. Methane emission estimates using chamber and tracer release experiments for a municipal waste water treatment plant. Atmos. Meas. Tech. 2015, 8 (7), 2853−2867. (34) Sun, J.; Hu, S.; Sharma, K. R.; Bustamante, H.; Yuan, Z. Impact of reduced water consumption on sulfide and methane production in rising main sewers. J. Environ. Manage. 2015, 154, 307−315. (35) Beaulieu, J. J.; Shuster, W. D.; Rebholz, J. A. Nitrous oxide emissions from a large, impounded River: The Ohio River. Environ. Sci. Technol. 2010, 44 (19), 7527−7533. (36) Richey, J. E.; Devol, A. H.; Wofsy, S. C.; Victoria, R.; Riberio, M. N. G. Biogenic gases and the oxidation and reduction of carbon in Amazon River and floodplain waters. Limnol. Oceanogr. 1988, 33 (4), 551−561. (37) Borges, A. V.; Darchambeau, F.; Teodoru, C. R.; Marwick, T. R.; Tamooh, F.; Geeraert, N.; Omengo, F. O.; Guerin, F.; Lambert, T.; Morana, C.; Okuku, E.; Bouillon, S. Globally significant greenhousegas emissions from African inland waters. Nat. Geosci. 2015, 8 (8), 637−642. (38) BMU. Die Wasserrahmenrichtlinie - Auf dem Weg zu guten Gewässern - Ergebnisse der Bewirtschaftungsplanung 2009 in Deutschland; Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit Umweltbundesamt: Bonn, 2010. (39) http://www.bmub.bund.de/en/topics/water-waste-soil/watermanagement/wastewater/.

I


Export of Dissolved Methane and Carbon Dioxide with Effluents from

Recommend Documents